Fault tolerant voice processing system

1. A fault tolerance call processing system for connection to the public switched telephone network comprising: a plurality of client computers containing call processing applications; a plurality of voice processing resources required for processing calls; and at least two computer telephony servers for interfacing between said client computers and said voice processing resources, each of said servers containing one or more processors communicating with said call processing applications and said resources, each of said processors having one or more primary and/or standby functional components each of which carrying out a particular call processing function, each of said standby functional components beginning to work only when its counterpart primary functional component stops operating whereby a fault tolerance at the server level is achieved by shadowing between said primary functional components and their counterpart standby functional components; wherein each of said two servers contains at least one of said primary functional components and at least one of said standby functional components, and at least one of said primary functional components and its standby functional components are located in different servers.

2. The system of claim 1 wherein each processor only contains one of the functional components.

3. The system of claim 1 at least one of said processors contains at least one of said primary functional components and at least one of said standby functional components.

4. The system of claim 1 wherein each of said primary components and its standby counterpart are located in different processors.

5. The system of claim 1 wherein each of said primary functional components has a standby functional component.

6. A two-tier fault tolerance call processing system for connection to the public switched telephone network comprising: a plurality of platforms, each of said platforms containing one or more circuit boards, and each of said circuit boards containing at least one call processing resources; one or more computer telephony servers for interfacing to a plurality, of client computers containing call processing applications, each of said servers containing one or more processors communicating with said call processing applications and each of said platforms, each of said processors having one or more primary and/or standby functional components each of which carrying out a particular call processing function; allocation software running on said servers for allocating, in responses to messages from said call processing applications, a group of resources from said platforms for a particular telephone call being processed, said allocation software being arranged to allocate the resources either from the same or different platforms, depending upon availability; wherein at least one of said primary functional components has a standby functional component on said servers, said standby functional component beginning to work only when its counterpart primary functional component stops operating whereby a fault tolerance at the server level is achieved by shadowing between said primary functional components and their counterpart standby functional components; and wherein a fault tolerance among the resources is achieved by re-allocating said group of the resources among operable resources on all said platforms when a resource stops operating.

7. The system of claim 6 wherein there are at least two servers and each of said primary functional components and its standby counterpart are accommodated by different servers.

8. The system of claim 7 wherein at least one of said servers contains at least one of said primary functional components and at least one of said standby functional components.

9. The system of claim 8 wherein each of said servers accommodates either only said primary functional components or only said standby functional components.

10. The system of claim 6 wherein said allocation software has a priority scheme to allocate resources to the extent that such allocation permits optimal call processing.

11. The system of claim 6 wherein said servers and said platforms communicate using an S.300 protocol.

12. The system of claim 6 wherein said sources are from different vendors.

13. The system of claim 6 wherein said allocation software is configured to route information for processing a call to and from resources allocated for that call such that the call processing application processing that call processes the call in a substantially identical manner whether the allocated resources are from the same or different boards or platforms, or from different vendors.

14. The system of claim 6 wherein at least one of the server includes a user interface and software which allow a user to specify resources to be used in processing a call, thereby overriding the resources allocated by the server.

15. A method of achieving fault tolerance in a call processing system, said call processing system comprising a plurality of call processing resources and at least two computer telephone servers, each of said servers containing one or more processors communicating with call processing applications and each of said resources, at least one of said processors having one or more primary functional components each of which carrying out a particular call processing function, the method comprising the steps of: providing on said processors of said servers standby functional components for said primary functional components such that each of said two servers contains at least one of said primary functional components and at least one of said standby functional components, and at least one of said primary functional components and its standby functional components are located in different servers; and keeping each of said standby functional components shadowed when its counterpart primary component operates properly and using said shadowed standby functional component when said primary component stops operating.

16. The method of claim 15 wherein said primary functional components and their standby functional components are arranged such that each of said primary components and its counterpart standby component are not on the same server.

17. The method of claim 15 wherein each of said primary components has a standby component.

18. The method of claim 15 wherein each of said processors contains only one of said functional components.

19. The method of claim 15 wherein at least one of said processors contains at least one of said primary components and one of said standby components.

20. The method of claim 19 wherein each of said primary components and its standby component are located on different processors.

21. A method of achieving two-tier fault tolerance in a call processing system, said call processing system comprising a plurality of platforms containing call processing resources and one or more computer telephone servers for communicating with call processing applications and said platforms, each of said servers containing one or more processors communicating with said call processing applications and each of said platforms, each of said processors having one or more primary and/or standby functional components each of which carrying out a particular call processing function, the method comprising the steps of: providing on said processors of said servers a standby functional component for at least one of said primary functional components; keeping said standby functional component shadowed when its counterpart primary component operates properly and using said shadowed standby functional component when said primary component stops operating; and redirecting the load of the system among the available resources when a resource stops operating.

22. The method of claim 21 wherein said redirection of the load is performed according to a priority scheme.

23. The method of claim 21 wherein said primary functional components and their standby functional components are arranged such that each of said primary components and its counterpart standby component are not on the same server.

24. The method of claim 23 wherein each of said servers accommodates some of said primary components and some of said standby components such that the load of the system are shared among all of said servers.